WO2022054231A1 - Transmission/reception apparatus, transmitter, signal generation method, and signal generation program - Google Patents

Transmission/reception apparatus, transmitter, signal generation method, and signal generation program Download PDF

Info

Publication number
WO2022054231A1
WO2022054231A1 PCT/JP2020/034492 JP2020034492W WO2022054231A1 WO 2022054231 A1 WO2022054231 A1 WO 2022054231A1 JP 2020034492 W JP2020034492 W JP 2020034492W WO 2022054231 A1 WO2022054231 A1 WO 2022054231A1
Authority
WO
WIPO (PCT)
Prior art keywords
parameters
multilevel
transmission
signals
learning
Prior art date
Application number
PCT/JP2020/034492
Other languages
French (fr)
Inventor
Nikolaos-Panteleimon DIAMANTOPOULOS
Shinji Matsuo
Original Assignee
Nippon Telegraph And Telephone Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nippon Telegraph And Telephone Corporation filed Critical Nippon Telegraph And Telephone Corporation
Priority to JP2023512729A priority Critical patent/JP7480912B2/en
Priority to PCT/JP2020/034492 priority patent/WO2022054231A1/en
Publication of WO2022054231A1 publication Critical patent/WO2022054231A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/32—Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34—Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems

Definitions

  • the present invention relates to a transmission/reception apparatus, a signal generation method, and a signal generation program, which are applied to multi-carrier transmission.
  • multilevel modulation methods such as QPSK (Quadrature Phase Shift Keying) and 16-QAM (Quadrature Amplitude Modulation), which convert phase information or intensity information into multilevel data, show great promise for construction of an optical transmission system with a high frequency efficiency.
  • QPSK Quadrature Phase Shift Keying
  • 16-QAM Quadrature Amplitude Modulation
  • a single-carrier modulation method When transmitting a QAM signal, a single-carrier modulation method can improve the signal speed and the spectrum efficiency and increase the transmission capacity. On the other hand, the single-carrier modulation method needs a high SN ratio (Signal-to-Noise Ratio), and the transmission distance is shortened. A change in the SN ratio depending on the frequency also limits the transmission capacity.
  • the multi-carrier transmission method is more preferable than the single-carrier modulation method.
  • QAM symbols having different information entropies are assigned to different frequency sub-bands.
  • Probabilistic shaping is one of coding methods, and this can improve an SN ratio characteristic by changing the probability in a symbol of a signal point arrangement (constellation) of multilevel modulation. For example, the occurrence probability of a symbol with a small amplitude is made higher than the occurrence probability of a symbol with a large amplitude, thereby improving the frequency use efficiency.
  • Fig. 6A shows an example of a 64-QAM constellation (signal point arrangement diagram) by probabilistic shaping.
  • the SN ratio is 20 dB, and the entropy is obtained as 4.8906 bits/symbol under the environment of Additive White Gaussian Noise (to be referred to as "AWGN” hereinafter).
  • AWGN Additive White Gaussian Noise
  • a modulation method of assigning an entropy (information amount) to a sub-carrier in multi-carrier modulation based on the probabilistic shaping is entropy loading.
  • entropy loading as shown in Fig. 6B, a different sub-carrier or sub-band is modulated by PS-QAM having an information entropy based on an SN ratio distribution.
  • an entropy profile is assigned to a different sub-carrier or sub-band. At this time, it is necessary to efficiently, easily, and correctly assign the entropy.
  • the conventional entropy loading cannot be performed for a lot of sub-carriers, and an enormous time, labor, and cost are needed to assign entropies to a lot of sub-carriers.
  • a multilevel signal generation method of generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation wherein initial values of parameters of probability distributions used in an adaptive learning algorithm are decided based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, the adaptive learning algorithm is executed based on a unified performance metric, and the entropies are decided.
  • a multilevel signal generation method of generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation comprising: the step of evaluating SN ratios by comparing learning multilevel signals at the time of transmission and reception; the step of deciding, based on the SN ratios, learning initial values of parameters of probability distributions for determining the entropies; the step of calculating a unified performance metric by an adaptive learning algorithm using the learning initial values; the step of comparing the unified performance metric and a target value of the unified performance metric; and the step of, if a difference between the unified performance metric and the target value is larger than a predetermined range, newly calculating the parameters and calculating the unified performance metric using the parameters.
  • a transmission/reception apparatus comprising a transmitter, a receiver, and a communication channel, wherein the receiver evaluates SN ratios of learning multilevel signals, the transmitter decides, based on the SN ratios, learning initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier, the parameters are calculated by an adaptive learning algorithm using the learning initial values, and multilevel signals are generated and transmitted using the parameters.
  • a transmission/reception apparatus comprising a transmitter, a receiver, and a communication channel, characterized in that the transmitter comprises a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier, and a modulator that generates and transmits multilevel signals using the parameters, the receiver comprises a demodulator that receives and demodulates the multilevel signals, and a reception controller that evaluates SN ratios of learning multilevel signals, and the transmission controller decides learning initial values of the parameters based on the SN ratios, and calculates the parameters by an adaptive learning algorithm using the learning initial values.
  • a transmitter that transmits multilevel signals to a receiver, characterized by comprising: a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier; and a modulator that generates and transmits multilevel signals using the parameters, wherein the transmission controller decides learning initial values of the parameters based on SN ratios of learning multilevel signals evaluated by the receiver, and calculates the parameters by an adaptive learning algorithm using the learning initial values.
  • a multilevel signal generation program configured to cause a transmitter to function, characterized by causing the transmitter that transmits multilevel signals to a receiver to: decide initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier, which are used in an adaptive learning algorithm, based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, and execute processing of executing the adaptive learning algorithm and calculating entropies.
  • a transmission/reception apparatus a transmitter, a signal generation method, and a signal generation program, which can assign entropies to a large set of sub-carriers and efficiently generate and transmit/receive entropy-loaded multi-carrier PS-multilevel signals such as PS-QAM signals in a multi-carrier modulation format.
  • Fig. 1 is a block diagram showing the arrangement of a transmission/reception apparatus according to the first embodiment of the present invention
  • Fig. 2 is a flowchart for explaining a signal generation method according to the first embodiment of the present invention
  • Fig. 3A is a graph showing a simulation result in the first example of the present invention
  • Fig. 3B is a graph showing a simulation result in the first example of the present invention
  • Fig. 4 is a graph showing a simulation result in the first example of the present invention
  • Fig. 5 is a block diagram showing an example of the arrangement of a computer according to an embodiment of the present invention
  • Fig. 6A is a view showing an example of a QAM constellation (signal point arrangement diagram) of a conventional method
  • Fig. 6B is a view showing entropy loading in multi-carrier transmission of the conventional method.
  • Fig. 1 shows an example of the arrangement of a transmission/reception apparatus 1.
  • the transmission/reception apparatus 1 includes a transmitter 11, a receiver 12, and a communication channel 131 that connects the transmitter 11 and the receiver 12.
  • the transmitter 11 includes an information source coder 111, a communication channel coder 112, and a modulator 113.
  • the information source coder 111 executes information source coding (data compression) and compresses the data of an information source to more efficiently transmit data.
  • the communication channel coder 112 executes communication channel coding (error detection and correction), and adds a data bit (redundant bit) to strengthen resistance to a failure such as noise existing on the communication channel.
  • the modulator 113 modulates a laser beam by a modulation signal (electric signal), thereby generating an optical signal.
  • a directly modulated laser is used.
  • a DFB (Distributed FeedBack) semiconductor laser with a DBR (Distributed Bragg Reflector) region is used (for example, N.P. Diamantopoulos, et al., "Net 321.24-Gb/s IMDD Transmission Based on a >100-GHz Bandwidth Directly-Modulated Laser," in Proc. Optical Networking and Communication Conference & Exhibition (OFC) 2020, San Diego, CA, USA, 8-12 March, 2020, paper Th4C.1.).
  • the active layer is made of an InP-based semiconductor, and the oscillation wavelength is the 1.55-micrometer band.
  • the modulator 113 may use another wavelength band, or may be made of another material.
  • a DFB laser, a DBR laser, or the like may be used as the directly modulated laser.
  • an EA-DFB laser or an external modulation device such as a combination of a semiconductor laser and an MZ modulator may be used.
  • a signal propagates through optical fiber that is used as the communication channel 131, and is received by the receiver 12.
  • the receiver 12 includes a demodulator 121, a communication channel decoder 122, and an information source decoder 123.
  • the demodulator 121 demodulates a received signal.
  • the demodulated received signal is input to the communication channel decoder 122, and decoding corresponding to the communication channel coding is performed.
  • the information source decoder 123 performs decoding corresponding to the information source coding.
  • the transmitter 11 further includes a transmission control unit (controller) 114 and a transmission storage unit (memory) 115.
  • the receiver 12 further includes a reception control unit (controller) 124 and a reception storage unit (memory) 125.
  • an adaptive learning algorithm (to be described later) for signal generation is executed.
  • the reception control unit 124 performs SN ratio evaluation necessary for the adaptive learning algorithm and responds to the transmission control unit 114.
  • the reception storage unit 125 stores received signal data, an SN ratio evaluation program, and parameters.
  • the transmission control unit 114 decides parameters necessary for signal generation by probabilistic shaping based on the evaluated SN ratio. Based on the parameters, the modulator 113 generates signals and transmits them. In addition, the transmission storage unit 115 stores a program and parameters to be used in the adaptive learning algorithm.
  • the transmitter 11 in the transmission/reception apparatus 1 responds to the receiver 12 and executes the adaptive learning algorithm, thereby improving the throughput of optical signal transmission.
  • the modulator 113 assigns PS-QAMs to S subcarriers each having a bandwidth B i based on modulation orders M i and entropies H(X i ).
  • the entropy H(X i ) is given by
  • X i is a set of all possible symbols for the ith subcarrier.
  • P i defines a probability distribution.
  • the entropy is represented by bits/symbol.
  • PS-QAM The probability distribution in PS-QAM, that is, a transmission probability P i (x) of a symbol with an amplitude x is represented, based on Maxwell Boltzmann (to be referred to as "MB" hereinafter) statistics, by
  • a parameter v i is an MB constant that defines the PS distribution in each subcarrier.
  • the parameter v i represents a probability distribution that defines an entropy to be assigned to each subcarrier.
  • v is a vector formed by all values v i , R s,i is a symbol rate per subcarrier band, r mod is a coding rate that defines all overheads necessary for synchronization, equalization, and modulation, and r FEC is an FEC coding rate.
  • NGMI normalized generalized mutual information
  • the NGMI is a communication capacity that can be implemented when an ideal binary soft decision error correction code is used.
  • the NGMI shows the similarity between PS-QAMs transmitted from the transmitter (Tx) 11 and PS-QAMs received by the receiver (Rx) 12. The higher the NGMI is, the higher the SN ratio is.
  • n 1, 2,..., N.
  • x i ) is a conditional probability that defines an auxiliary channel between an input signal x i and an output signal y i , and is given by expression (5) under an AWGN environment.
  • NGMI is represented by expression (6) using a bitwise log-likelihood ratio (to be referred to as "LLR" hereinafter).
  • the LLR is an output obtained by demodulation using a soft decision demodulation method, and represents a likelihood of a coding bit 0 or 1. In general, if the LLR is a positive value, and its absolute value is large, the possibility that the coding bit is 1 is high. If the LLR is a negative value, and its absolute value is large, the possibility that the coding bit is 0 is high.
  • the Newton's method is used in executing the adaptive learning algorithm.
  • the partial derivative coefficients of the MB distribution, the entropy, the net data rate, and the NGMI are defined by expressions (7) to (12).
  • j an iteration index
  • Fig. 2 is a flowchart for explaining a signal generation method.
  • the SN ratio is evaluated to associate the SN ratio with M i and v i (step 22).
  • the receiver 12 compares the transmitted QAM symbols and the received QAM symbols, thereby evaluating the SN ratio in each subcarrier using a normal SN ratio evaluation method. The relationship between M i and v i at the time of transmission and the SN ratio in the receiver 12 is thus acquired.
  • an NGMI target (target value) or a NetRate target (target value) is decided. These values indicate a margin to a variation in noise in signal transmission.
  • the NGMI target is set to 0.86.
  • the NetRate target is set to 50 Gbit/s or 100 Gbit/s.
  • NGMI or NetRate as the calculation result and the NGMI target (NGMI*) or the NetRate target (NetRate*) are compared and determined (step 25).
  • v i is newly decided using equation (13) or (14), and the simulation is executed again from step 24 (step 26).
  • the determination criterion may be not only whether the NGMI (or NetRate) as the calculation result equals NGMI* (or NetRate*) but also whether the difference between them falls within a predetermined range.
  • NGMI and NGMI* are 0.005 or less, it may be determined that the performance criterion is reached. In this case, if the difference between NGMI and NGMI* is larger than 0.005, v i is newly decided, and the simulation is executed again from step 24 (step 26).
  • PS-QAM signals obtained by the simulation are transmitted (step 28).
  • the signal generation method executes the adaptive learning algorithm by using the unified performance metric, thereby optimizing the probability distributions over a large set of subcarriers in a multi-carrier modulation.
  • signals of which probabilistic shaping is performed based on the adaptive learning algorithm are generated in the arrangement of an optical transmission/reception apparatus 1 according to the first embodiment.
  • the Newton's method is used in the adaptive learning algorithm.
  • an RC filter with a 3-dB bandwidth of 80 GHz is used in consideration of a frequency dependency.
  • Fig. 4 shows a net data rate with a target NGMI for different SN ratios.
  • the system according to this embodiment has at least a 3-dB bandwidth of 80 GHz.
  • a net data rate of 400 Gb/s can be obtained for an average SN ratio of 17 dB or more in this system.
  • the parameter v i necessary for signal generation can be decided by the adaptive learning algorithm without repetitively performing comparison with a lookup table at the time of shaping of PS-QAMs.
  • the Newton's method is used to acquire an optimum solution in the adaptive learning algorithm.
  • an appropriate value that is, a value close to the optimum solution needs to be set as a learning initial value.
  • the learning initial values of M i and v i used in the adaptive learning algorithm are decided by SN ratio evaluation (steps 22 and 23), thereby appropriately setting the values.
  • NGMI or NetRate quickly converges to obtain an optimum solution, and PS-QAM signals can be generated.
  • the second embodiment of the present invention will be described next.
  • the arrangement of a transmission/reception apparatus according to the second embodiment is the same as in the first embodiment.
  • a least mean square method is used in place of the Newton's method in the signal generation method according to the first embodiment.
  • equations (15) and (16) are used in place of equations (13) and (14) in the adaptive algorithm according to the first embodiment.
  • the least mean square method is used to acquire an optimum solution in the adaptive learning algorithm.
  • an appropriate value that is, a value close to the optimum solution needs to be set as a learning initial value.
  • the learning initial values of M i and v i used in the adaptive learning algorithm are decided by SN ratio evaluation (steps 22 and 23), thereby appropriately setting the values.
  • NGMI or NetRate quickly converges to obtain an optimum solution, and PS-QAM signals can be generated.
  • Fig. 5 shows an example of the arrangement of the computer of the transmission control unit (controller)/transmission storage unit (memory) and the reception control unit (controller)/reception storage unit (memory) of the transmission/reception apparatus according to the embodiment of the present invention.
  • Each of the transmission control unit/transmission storage unit and the reception control unit/reception storage unit can be implemented by a computer 5 including a CPU (Central Processing Unit) 53, a storage device (storage unit) 52, and an interface device 51, and a program that controls these hardware resources.
  • a modulator, a demodulator, or the like is connected to the interface device 51.
  • the CPU 53 executes processing according to the embodiment of the present invention in accordance with a signal generation program stored in the storage device (storage unit) 52.
  • the signal generation program thus makes the transmission/reception apparatus function.
  • the computer may be provided in the apparatus, or at least some of the functions of the computer may be implemented using an external computer.
  • a storage medium 54 outside the apparatus may be used, and a signal generation program stored in the storage medium 54 may be read out and executed.
  • the storage medium 54 includes various kinds of magnetic recording media, a magneto-optical recording medium, a CD-ROM, a CD-R, and various kinds of memories.
  • the signal generation program may be supplied to the computer via a communication network such as the Internet.
  • 16-QAM signals are used as multilevel signals.
  • the present invention is not limited to this.
  • Other multilevel signals such as PSK signals, 4-QAM signals, or 64-QAM signals can also be used.
  • the present invention can be applied to an optical transmission system, devices such as a transmitter and a receiver in the optical transmission system, a modulator that generates an optical signal, and the like.
  • 1...transmission/reception apparatus 11...transmitter, 111...information source coder, 112...communication channel coder, 113...modulator, 114...transmission control unit (controller), 115...transmission storage unit (memory), 12...receiver, 121...demodulator, 122...communication channel decoder, 123...information source decoder, 124...reception control unit, (controller), 125...reception storage unit (memory), 131...communication channel

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
  • Optical Communication System (AREA)

Abstract

A multilevel signal generation method of this invention is a method of generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation, wherein initial values of parameters of probability distributions used in an adaptive learning algorithm are decided based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, the adaptive learning algorithm is executed based on a unified performance metric, and the entropies are decided. Hence, this invention can provide a signal generation method of efficiently generating and transmitting/receiving entropy-loaded multi-carrier PS-multilevel signals such as PS-QAM signals in a multi-carrier modulation format.

Description

TRANSMISSION/RECEPTION APPARATUS, TRANSMITTER, SIGNAL GENERATION METHOD, AND SIGNAL GENERATION PROGRAM
The present invention relates to a transmission/reception apparatus, a signal generation method, and a signal generation program, which are applied to multi-carrier transmission.
In recent years, to implement an optical transmission system with a large capacity and a wide band, research and development on advanced multilevel modulation and waveform shaping has progressed. In particular, multilevel modulation methods such as QPSK (Quadrature Phase Shift Keying) and 16-QAM (Quadrature Amplitude Modulation), which convert phase information or intensity information into multilevel data, show great promise for construction of an optical transmission system with a high frequency efficiency.
When transmitting a QAM signal, a single-carrier modulation method can improve the signal speed and the spectrum efficiency and increase the transmission capacity. On the other hand, the single-carrier modulation method needs a high SN ratio (Signal-to-Noise Ratio), and the transmission distance is shortened. A change in the SN ratio depending on the frequency also limits the transmission capacity.
For this reason, to maximize throughput in optical transmission, the multi-carrier transmission method is more preferable than the single-carrier modulation method. In the multi-carrier transmission method, QAM symbols having different information entropies are assigned to different frequency sub-bands.
Probabilistic shaping (PS) is one of coding methods, and this can improve an SN ratio characteristic by changing the probability in a symbol of a signal point arrangement (constellation) of multilevel modulation. For example, the occurrence probability of a symbol with a small amplitude is made higher than the occurrence probability of a symbol with a large amplitude, thereby improving the frequency use efficiency.
Fig. 6A shows an example of a 64-QAM constellation (signal point arrangement diagram) by probabilistic shaping. The SN ratio is 20 dB, and the entropy is obtained as 4.8906 bits/symbol under the environment of Additive White Gaussian Noise (to be referred to as "AWGN" hereinafter).
A modulation method of assigning an entropy (information amount) to a sub-carrier in multi-carrier modulation based on the probabilistic shaping is entropy loading.
In entropy loading, as shown in Fig. 6B, a different sub-carrier or sub-band is modulated by PS-QAM having an information entropy based on an SN ratio distribution.
More specifically, to maximize throughput or minimize an error characteristic, an entropy profile is assigned to a different sub-carrier or sub-band. At this time, it is necessary to efficiently, easily, and correctly assign the entropy.
Conventionally, the entropy assignment has been done manually or using a three-dimensional lookup table. (non-patent literature 1).
Di Che and W. Shieh, "Squeezing out the last few bits from band-limited channels with entropy loading," Opt. Express, vol. 27, no. 7, pp. 9321-9329, Apr. 2019.
However, when manually assigning an entropy, it is impossible to do assignment to a lot of sub-carriers. When using a lookup table, large-scale simulations are needed to construct a three-dimensional matrix first in the lookup table.
As described above, the conventional entropy loading cannot be performed for a lot of sub-carriers, and an enormous time, labor, and cost are needed to assign entropies to a lot of sub-carriers.
In order to solve the above-described problem, according to the present invention, there is provided a multilevel signal generation method of generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation, wherein initial values of parameters of probability distributions used in an adaptive learning algorithm are decided based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, the adaptive learning algorithm is executed based on a unified performance metric, and the entropies are decided.
According to the present invention, there is also provided a multilevel signal generation method of generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation, comprising: the step of evaluating SN ratios by comparing learning multilevel signals at the time of transmission and reception; the step of deciding, based on the SN ratios, learning initial values of parameters of probability distributions for determining the entropies; the step of calculating a unified performance metric by an adaptive learning algorithm using the learning initial values; the step of comparing the unified performance metric and a target value of the unified performance metric; and the step of, if a difference between the unified performance metric and the target value is larger than a predetermined range, newly calculating the parameters and calculating the unified performance metric using the parameters.
According to the present invention, there is also provided a transmission/reception apparatus comprising a transmitter, a receiver, and a communication channel, wherein the receiver evaluates SN ratios of learning multilevel signals, the transmitter decides, based on the SN ratios, learning initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier, the parameters are calculated by an adaptive learning algorithm using the learning initial values, and multilevel signals are generated and transmitted using the parameters.
According to the present invention, there is also provided a transmission/reception apparatus comprising a transmitter, a receiver, and a communication channel, characterized in that the transmitter comprises a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier, and a modulator that generates and transmits multilevel signals using the parameters, the receiver comprises a demodulator that receives and demodulates the multilevel signals, and a reception controller that evaluates SN ratios of learning multilevel signals, and the transmission controller decides learning initial values of the parameters based on the SN ratios, and calculates the parameters by an adaptive learning algorithm using the learning initial values.
According to the present invention, there is also provided a transmitter that transmits multilevel signals to a receiver, characterized by comprising: a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier; and a modulator that generates and transmits multilevel signals using the parameters, wherein the transmission controller decides learning initial values of the parameters based on SN ratios of learning multilevel signals evaluated by the receiver, and calculates the parameters by an adaptive learning algorithm using the learning initial values.
According to the present invention, there is also provided a multilevel signal generation program configured to cause a transmitter to function, characterized by causing the transmitter that transmits multilevel signals to a receiver to: decide initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier, which are used in an adaptive learning algorithm, based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, and execute processing of executing the adaptive learning algorithm and calculating entropies.
According to the present invention, it is possible to provide a transmission/reception apparatus, a transmitter, a signal generation method, and a signal generation program, which can assign entropies to a large set of sub-carriers and efficiently generate and transmit/receive entropy-loaded multi-carrier PS-multilevel signals such as PS-QAM signals in a multi-carrier modulation format.
Fig. 1 is a block diagram showing the arrangement of a transmission/reception apparatus according to the first embodiment of the present invention; Fig. 2 is a flowchart for explaining a signal generation method according to the first embodiment of the present invention; Fig. 3A is a graph showing a simulation result in the first example of the present invention; Fig. 3B is a graph showing a simulation result in the first example of the present invention; Fig. 4 is a graph showing a simulation result in the first example of the present invention; Fig. 5 is a block diagram showing an example of the arrangement of a computer according to an embodiment of the present invention; Fig. 6A is a view showing an example of a QAM constellation (signal point arrangement diagram) of a conventional method; and Fig. 6B is a view showing entropy loading in multi-carrier transmission of the conventional method.
(First Embodiment)
The first embodiment of the present invention will be described with reference to Figs. 1 and 2.
(Arrangement of Transmission/Reception Apparatus)
Fig. 1 shows an example of the arrangement of a transmission/reception apparatus 1. The transmission/reception apparatus 1 includes a transmitter 11, a receiver 12, and a communication channel 131 that connects the transmitter 11 and the receiver 12.
The transmitter 11 includes an information source coder 111, a communication channel coder 112, and a modulator 113.
The information source coder 111 executes information source coding (data compression) and compresses the data of an information source to more efficiently transmit data.
The communication channel coder 112 executes communication channel coding (error detection and correction), and adds a data bit (redundant bit) to strengthen resistance to a failure such as noise existing on the communication channel.
The modulator 113 modulates a laser beam by a modulation signal (electric signal), thereby generating an optical signal. As the modulator 113, a directly modulated laser is used. As the directly modulated laser, a DFB (Distributed FeedBack) semiconductor laser with a DBR (Distributed Bragg Reflector) region is used (for example, N.P. Diamantopoulos, et al., "Net 321.24-Gb/s IMDD Transmission Based on a >100-GHz Bandwidth Directly-Modulated Laser," in Proc. Optical Networking and Communication Conference & Exhibition (OFC) 2020, San Diego, CA, USA, 8-12 March, 2020, paper Th4C.1.). The active layer is made of an InP-based semiconductor, and the oscillation wavelength is the 1.55-micrometer band. The modulator 113 may use another wavelength band, or may be made of another material. As the directly modulated laser, a DFB laser, a DBR laser, or the like may be used.
As the modulator 113, an EA-DFB laser or an external modulation device such as a combination of a semiconductor laser and an MZ modulator may be used.
A signal propagates through optical fiber that is used as the communication channel 131, and is received by the receiver 12.
The receiver 12 includes a demodulator 121, a communication channel decoder 122, and an information source decoder 123. The demodulator 121 demodulates a received signal. The demodulated received signal is input to the communication channel decoder 122, and decoding corresponding to the communication channel coding is performed. The information source decoder 123 performs decoding corresponding to the information source coding.
The transmitter 11 further includes a transmission control unit (controller) 114 and a transmission storage unit (memory) 115. The receiver 12 further includes a reception control unit (controller) 124 and a reception storage unit (memory) 125.
In a response between the transmission control unit 114 and the reception control unit 124, an adaptive learning algorithm (to be described later) for signal generation is executed. The reception control unit 124 performs SN ratio evaluation necessary for the adaptive learning algorithm and responds to the transmission control unit 114. The reception storage unit 125 stores received signal data, an SN ratio evaluation program, and parameters.
The transmission control unit 114 decides parameters necessary for signal generation by probabilistic shaping based on the evaluated SN ratio. Based on the parameters, the modulator 113 generates signals and transmits them. In addition, the transmission storage unit 115 stores a program and parameters to be used in the adaptive learning algorithm.
The transmitter 11 in the transmission/reception apparatus 1 according to this embodiment responds to the receiver 12 and executes the adaptive learning algorithm, thereby improving the throughput of optical signal transmission.
(Principle of Signal Generation Method)
The modulator 113 assigns PS-QAMs to S subcarriers each having a bandwidth Bi based on modulation orders Mi and entropies H(Xi). The entropy H(Xi) is given by
Figure JPOXMLDOC01-appb-M000001
Here, i is the index of a subcarrier (i = 1, 2,..., S).
Xi is a set of all possible symbols for the ith subcarrier. Pi defines a probability distribution. The entropy is represented by bits/symbol.
The probability distribution in PS-QAM, that is, a transmission probability Pi(x) of a symbol with an amplitude x is represented, based on Maxwell Boltzmann (to be referred to as "MB" hereinafter) statistics, by
Figure JPOXMLDOC01-appb-M000002
Here, a parameter vi is an MB constant that defines the PS distribution in each subcarrier. In other words, the parameter vi represents a probability distribution that defines an entropy to be assigned to each subcarrier.
In this system, a net data rate after all forward error correction (to be referred to as "FEC" hereinafter) codes are removed is represented by
Figure JPOXMLDOC01-appb-M000003
Here, v is a vector formed by all values vi, Rs,i is a symbol rate per subcarrier band, rmod is a coding rate that defines all overheads necessary for synchronization, equalization, and modulation, and rFEC is an FEC coding rate.
In this embodiment, as a unified performance metric representing transmission performance in multi-carrier PS modulation, normalized generalized mutual information (to be referred to as "NGMI" hereinafter) represented by expression (4) is used.
The NGMI is a communication capacity that can be implemented when an ideal binary soft decision error correction code is used. In this embodiment, the NGMI shows the similarity between PS-QAMs transmitted from the transmitter (Tx) 11 and PS-QAMs received by the receiver (Rx) 12. The higher the NGMI is, the higher the SN ratio is.
Figure JPOXMLDOC01-appb-M000004
Here, mi = log2(Mi), and n = 1, 2,..., N.
qi(yi|xi) is a conditional probability that defines an auxiliary channel between an input signal xi and an output signal yi, and is given by expression (5) under an AWGN environment.
Figure JPOXMLDOC01-appb-M000005
Also, NGMI is represented by expression (6) using a bitwise log-likelihood ratio (to be referred to as "LLR" hereinafter).
Figure JPOXMLDOC01-appb-M000006
The LLR is an output obtained by demodulation using a soft decision demodulation method, and represents a likelihood of a coding bit 0 or 1. In general, if the LLR is a positive value, and its absolute value is large, the possibility that the coding bit is 1 is high. If the LLR is a negative value, and its absolute value is large, the possibility that the coding bit is 0 is high.
Figure JPOXMLDOC01-appb-I000007
In this embodiment, the Newton's method is used in executing the adaptive learning algorithm. Hence, concerning vi or v, the partial derivative coefficients of the MB distribution, the entropy, the net data rate, and the NGMI are defined by expressions (7) to (12).
Figure JPOXMLDOC01-appb-M000008
Figure JPOXMLDOC01-appb-M000009
Figure JPOXMLDOC01-appb-M000010
Figure JPOXMLDOC01-appb-M000011
Figure JPOXMLDOC01-appb-M000012
Figure JPOXMLDOC01-appb-M000013
Finally, based on expressions (1) to (12), the relationship between vj+1 and vj is defined by equation (13) or (14). In equation (13), an NGMI target (NGMI*) is used.
Figure JPOXMLDOC01-appb-M000014
In equation (14), a target (NetRate*) of the net data rate is used.
Figure JPOXMLDOC01-appb-M000015
Here, j represents an iteration index.
(Procedure of Signal Generation Method)
The adaptive learning algorithm is executed using the above-described expressions. Fig. 2 is a flowchart for explaining a signal generation method.
First, S, Bi, rmod, and rrec that are input parameters are decided (step 21).
Next, the SN ratio is evaluated to associate the SN ratio with Mi and vi (step 22).
The modulator 113 in the transmitter 11 assigns learning QAMs to all subcarriers of discrete multi-tone (DMT), and transmits these to the demodulator 121 in the receiver (Rx) 12. For example, when QAM signals with Mi = 4, and vi = 0, that is, QAM signals that have not undergone probabilistic shaping is transmitted, the SN ratio in the receiver (Rx) 12 is evaluated. Here, the symbol of the modulation order Mi = 4 means 16-QAM.
The receiver 12 compares the transmitted QAM symbols and the received QAM symbols, thereby evaluating the SN ratio in each subcarrier using a normal SN ratio evaluation method. The relationship between Mi and vi at the time of transmission and the SN ratio in the receiver 12 is thus acquired.
Next, the initial values (to be referred to as "learning initial values" hereinafter) of Mi and vi to be used in a simulation of the adaptive learning algorithm are decided based on the relationship between Mi and vi and the SN ratio, which is obtained in step 22 (step 23). For example, if the SN ratio is lower than 5 dB, subcarriers are not used. If the SN ratio falls within the range of 5 dB (inclusive) to 7 dB (inclusive), Mi = 4 and vi = 0 are used. To decide the learning initial values, a lookup table that holds the relationship between Mi and vi and the SN ratio may be used.
Next, to determine the performance of signal transmission, transmission of PS-QAM signals obtained by the learning initial values of Mi and vi from the transmitter (Tx) to the receiver (Rx) is simulated (step 24).
As a reference used to determine the performance of signal transmission, an NGMI target (target value) or a NetRate target (target value) is decided. These values indicate a margin to a variation in noise in signal transmission.
For example, the NGMI target is set to 0.86. Alternatively, the NetRate target is set to 50 Gbit/s or 100 Gbit/s.
Next, the NGMI or NetRate, which is the metric of transmission performance, is calculated using expressions (3) to (6).
Next, the NGMI or NetRate as the calculation result and the NGMI target (NGMI*) or the NetRate target (NetRate*) are compared and determined (step 25).
If the NGMI or NetRate as the calculation result does not equal NGMI* or NetRate*, vi is newly decided using equation (13) or (14), and the simulation is executed again from step 24 (step 26).
If the NGMI or NetRate as the calculation result equals NGMI* or NetRate*, it is determined that the performance criterion is reached (performance target is achieved), and the simulation is ended (step 27).
Here, the determination criterion may be not only whether the NGMI (or NetRate) as the calculation result equals NGMI* (or NetRate*) but also whether the difference between them falls within a predetermined range.
For example, if the difference between NGMI and NGMI* is 0.005 or less, it may be determined that the performance criterion is reached. In this case, if the difference between NGMI and NGMI* is larger than 0.005, vi is newly decided, and the simulation is executed again from step 24 (step 26).
Finally, PS-QAM signals obtained by the simulation are transmitted (step 28).
The signal generation method according to this embodiment executes the adaptive learning algorithm by using the unified performance metric, thereby optimizing the probability distributions over a large set of subcarriers in a multi-carrier modulation.
(Example)
In this example, signals of which probabilistic shaping is performed based on the adaptive learning algorithm are generated in the arrangement of an optical transmission/reception apparatus 1 according to the first embodiment. The Newton's method is used in the adaptive learning algorithm.
In this example, an RC filter with a 3-dB bandwidth of 80 GHz is used in consideration of a frequency dependency. Discrete multi-tone (DMT) is used as multi-carrier modulation, the number of subcarriers is S = 512, and the sampling speed is 160 GHz.
As a result, the bandwidth of the subcarrier is B = Bi = 160/512 = 312.5 MHz, and the symbol speed is Rs = Rs,i = B/2 = 156.25.
A composite overhead for time synchronization and time equalization is about 1.9%, and this value corresponds to rmod = 0.9814. A code of concatenated FEC is added at a total coding rate rFEC = 0.826 (about 21% overhead), and the target NGMI is 0.857.
The adaptive learning algorithm is executed under this condition. Figs. 3A and 3B show simulation results by the adaptive algorithm. Net data rates for AWGN having an SN ratio = 12 dB (Fig. 3A) and AWGN having an SN ratio = 18 dB (Fig. 3B) were calculated.
As the net data rate, 250 Gb/s is obtained when SN ratio = 12 dB, and 400 Gb/s or more is obtained when SN ratio = 18 dB. At this time, it can be seen that the number of iterations necessary for sufficient convergence is less than six.
Fig. 4 shows a net data rate with a target NGMI for different SN ratios. As described above, the system according to this embodiment has at least a 3-dB bandwidth of 80 GHz. As is apparent, a net data rate of 400 Gb/s can be obtained for an average SN ratio of 17 dB or more in this system.
As described above, according to the signal generation method of this embodiment, the parameter vi necessary for signal generation can be decided by the adaptive learning algorithm without repetitively performing comparison with a lookup table at the time of shaping of PS-QAMs.
As a result, according to the transmission/reception apparatus of this embodiment, it is possible to quickly generate PS-QAM signals and implement improvement of the throughput of optical transmission.
In the signal generation method according to this embodiment, the Newton's method is used to acquire an optimum solution in the adaptive learning algorithm. Although the Newton's method can quickly obtain the optimum solution, an appropriate value, that is, a value close to the optimum solution needs to be set as a learning initial value.
In this embodiment, the learning initial values of Mi and vi used in the adaptive learning algorithm are decided by SN ratio evaluation (steps 22 and 23), thereby appropriately setting the values.
As a result, in the adaptive learning algorithm, by the Newton's method, NGMI or NetRate quickly converges to obtain an optimum solution, and PS-QAM signals can be generated.
(Second Embodiment)
The second embodiment of the present invention will be described next. The arrangement of a transmission/reception apparatus according to the second embodiment is the same as in the first embodiment. In a signal generation method according to this embodiment, a least mean square method is used in place of the Newton's method in the signal generation method according to the first embodiment.
More specifically, equations (15) and (16) are used in place of equations (13) and (14) in the adaptive algorithm according to the first embodiment.
In equation (15), an NGMI target (NGMI*) is used.
Figure JPOXMLDOC01-appb-M000016
In equation (16), a net data rate target (NetRate*) is used.
Figure JPOXMLDOC01-appb-M000017
In the signal generation method according to this embodiment, the least mean square method is used to acquire an optimum solution in the adaptive learning algorithm. Although the least mean square method can quickly obtain the optimum solution, an appropriate value, that is, a value close to the optimum solution needs to be set as a learning initial value.
In this embodiment, the learning initial values of Mi and vi used in the adaptive learning algorithm are decided by SN ratio evaluation (steps 22 and 23), thereby appropriately setting the values.
As a result, in the adaptive learning algorithm, by the least mean square method, NGMI or NetRate quickly converges to obtain an optimum solution, and PS-QAM signals can be generated.
Fig. 5 shows an example of the arrangement of the computer of the transmission control unit (controller)/transmission storage unit (memory) and the reception control unit (controller)/reception storage unit (memory) of the transmission/reception apparatus according to the embodiment of the present invention. Each of the transmission control unit/transmission storage unit and the reception control unit/reception storage unit can be implemented by a computer 5 including a CPU (Central Processing Unit) 53, a storage device (storage unit) 52, and an interface device 51, and a program that controls these hardware resources. Here, a modulator, a demodulator, or the like is connected to the interface device 51. The CPU 53 executes processing according to the embodiment of the present invention in accordance with a signal generation program stored in the storage device (storage unit) 52. The signal generation program thus makes the transmission/reception apparatus function.
In the transmission control unit/transmission storage unit and the reception control unit/reception storage unit according to the embodiment of the present invention, the computer may be provided in the apparatus, or at least some of the functions of the computer may be implemented using an external computer. As the storage unit, a storage medium 54 outside the apparatus may be used, and a signal generation program stored in the storage medium 54 may be read out and executed. The storage medium 54 includes various kinds of magnetic recording media, a magneto-optical recording medium, a CD-ROM, a CD-R, and various kinds of memories. The signal generation program may be supplied to the computer via a communication network such as the Internet.
In the embodiment of the present invention, an example in which 16-QAM signals are used as multilevel signals has been described. However, the present invention is not limited to this. Other multilevel signals such as PSK signals, 4-QAM signals, or 64-QAM signals can also be used.
In the embodiment of the present invention, examples of the arrangements of a transmission/reception apparatus, a signal generation method, and the like have been described. However, the present invention is not limited to this. It is necessary to only exhibit the functions of a transmission/reception apparatus, a signal generation method, and the like and obtain the effects.
The present invention can be applied to an optical transmission system, devices such as a transmitter and a receiver in the optical transmission system, a modulator that generates an optical signal, and the like.
1...transmission/reception apparatus, 11...transmitter, 111...information source coder, 112...communication channel coder, 113...modulator, 114...transmission control unit (controller), 115...transmission storage unit (memory), 12...receiver, 121...demodulator, 122...communication channel decoder, 123...information source decoder, 124...reception control unit, (controller), 125...reception storage unit (memory), 131...communication channel

Claims (8)

  1. A multilevel signal generation method of
    generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation,
    wherein initial values of parameters of probability distributions used in an adaptive learning algorithm are decided based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, the adaptive learning algorithm is executed based on a unified performance metric, and the entropies are decided.
  2. A multilevel signal generation method of
    generating multilevel and multi-carrier signals by assigning entropies for each subcarrier by probabilistic shaping in a multi-carrier modulation, comprising:
    the step of evaluating SN ratios by comparing learning multilevel signals at the time of transmission and reception;
    the step of deciding, based on the SN ratios, learning initial values of parameters of probability distributions for determining the entropies;
    the step of calculating a unified performance metric by an adaptive learning algorithm using the learning initial values;
    the step of comparing the unified performance metric and a target value of the unified performance metric; and
    the step of, if a difference between the unified performance metric and the target value is larger than a predetermined range, newly calculating the parameters and calculating the unified performance metric using the parameters.
  3. The multilevel signal generation method
    according to claims 1 and 2, characterized in that the unified performance metric is one of normalized generalized mutual information and a net data rate.
  4. The multilevel signal generation method
    according to any one of claims 1 to 3, characterized in that in the adaptive learning algorithm, an optimum solution is obtained by one of the Newton's method and a least mean square method.
  5. A transmission/reception apparatus
    comprising a transmitter, a receiver, and a communication channel,
    wherein the receiver evaluates SN ratios of learning multilevel signals,
    the transmitter decides, based on the SN ratios, learning initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier,
    the parameters are calculated by an adaptive learning algorithm using the learning initial values, and
    multilevel signals are generated and transmitted using the parameters.
  6. A transmission/reception apparatus
    comprising a transmitter, a receiver, and a communication channel, characterized in that
    the transmitter comprises a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier, and
    a modulator that generates and transmits multilevel signals using the parameters,
    the receiver comprises a demodulator that receives and demodulates the multilevel signals, and
    a reception controller that evaluates SN ratios of learning multilevel signals, and
    the transmission controller decides learning initial values of the parameters based on the SN ratios, and
    calculates the parameters by an adaptive learning algorithm using the learning initial values.
  7. A transmitter that transmits multilevel signals to a receiver,
    characterized by comprising:
    a transmission controller that decides parameters of probability distributions for determining entropies to be assigned for each subcarrier; and
    a modulator that generates and transmits multilevel signals using the parameters,
    wherein the transmission controller decides learning initial values of the parameters based on SN ratios of learning multilevel signals evaluated by the receiver, and
    calculates the parameters by an adaptive learning algorithm using the learning initial values.
  8. A multilevel signal generation program
    configured to cause a transmitter to function, characterized by causing the transmitter that transmits multilevel signals to a receiver to:
    decide initial values of parameters of probability distributions for determining entropies to be assigned for each subcarrier, which are used in an adaptive learning algorithm, based on SN ratios evaluated by comparing learning multilevel signals at the time of transmission and reception, and
    execute processing of executing the adaptive learning algorithm and calculating entropies.
PCT/JP2020/034492 2020-09-11 2020-09-11 Transmission/reception apparatus, transmitter, signal generation method, and signal generation program WO2022054231A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2023512729A JP7480912B2 (en) 2020-09-11 2020-09-11 Transmitting/receiving device, transmitter, signal generating method, and signal generating program
PCT/JP2020/034492 WO2022054231A1 (en) 2020-09-11 2020-09-11 Transmission/reception apparatus, transmitter, signal generation method, and signal generation program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/034492 WO2022054231A1 (en) 2020-09-11 2020-09-11 Transmission/reception apparatus, transmitter, signal generation method, and signal generation program

Publications (1)

Publication Number Publication Date
WO2022054231A1 true WO2022054231A1 (en) 2022-03-17

Family

ID=80631437

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/034492 WO2022054231A1 (en) 2020-09-11 2020-09-11 Transmission/reception apparatus, transmitter, signal generation method, and signal generation program

Country Status (2)

Country Link
JP (1) JP7480912B2 (en)
WO (1) WO2022054231A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115643141A (en) * 2022-10-17 2023-01-24 西南交通大学 Modulation format identification method of probability shaping coherent optical communication system based on signal nonlinear power change

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180269979A1 (en) * 2017-03-06 2018-09-20 Nec Laboratories America, Inc Constellation optimization based on generalized mutual information over a nonlinear optical channel
US20190109752A1 (en) * 2017-10-09 2019-04-11 Nec Laboratories America, Inc. Probabilistic Shaping for Arbitrary Modulation Formats
US20190215077A1 (en) * 2018-01-05 2019-07-11 Zte Corporation Probabilistically shaped multi-level pulse modulation with gray code mapping

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180269979A1 (en) * 2017-03-06 2018-09-20 Nec Laboratories America, Inc Constellation optimization based on generalized mutual information over a nonlinear optical channel
US20190109752A1 (en) * 2017-10-09 2019-04-11 Nec Laboratories America, Inc. Probabilistic Shaping for Arbitrary Modulation Formats
US20190215077A1 (en) * 2018-01-05 2019-07-11 Zte Corporation Probabilistically shaped multi-level pulse modulation with gray code mapping

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
CHE DI, SHIEH WILLIAM: "Squeezing out the last few bits from band-limited channels with entropy loading", OPTICS EXPRESS, vol. 27, no. 7, 1 April 2019 (2019-04-01), pages 9321, XP055910448, DOI: 10.1364/OE.27.009321 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115643141A (en) * 2022-10-17 2023-01-24 西南交通大学 Modulation format identification method of probability shaping coherent optical communication system based on signal nonlinear power change

Also Published As

Publication number Publication date
JP7480912B2 (en) 2024-05-10
JP2023538433A (en) 2023-09-07

Similar Documents

Publication Publication Date Title
US11864007B2 (en) Communication systems capable of receiving and processing data using unequally spaced and uniform quadrature amplitude modulated 64 point symbol constellations
RU2536371C2 (en) Determining wireless link quality based on received signals
CN111010255B (en) IM/DD transmission method and system based on polarization code coding
CN108259135B (en) Atmospheric turbulence weakening resistant polarization code construction method based on Gaussian approximation theory
CN111030779B (en) Method for optimizing non-rate code degree distribution under compressed transmission of cloud access network
KR100957459B1 (en) Per stream rate control using ??? decoding
WO2022054231A1 (en) Transmission/reception apparatus, transmitter, signal generation method, and signal generation program
KR102027828B1 (en) Method and apparatus for estimating channel information
US9088327B2 (en) Systems and methods for channel quality indicator (CQI) selection for non-deterministic receivers
Okamoto et al. Laser phase noise tolerance of probabilistically-shaped constellations
CN112039536B (en) Adaptive polarization code coding and decoding method based on orthogonal frequency division multiplexing technology
KR20090128061A (en) Apparatus and method for signal detecton in mimo system
JP2019092014A (en) Frame error rate prediction device, wireless communication device and wireless communication system
US12035151B2 (en) Methods of receiving data transmitted using non-uniform multidimensional constellation and code rate pairs
McKiernan Foundational Techniques for Wireless Communications: Channel Coding, Modulation, and Equalization
Ibrahimov Investigation of Noise Immunity Telecommunication Systems According to the Criterion Energy Efficiency
CN117579161A (en) Optical fiber nonlinear noise equalization method and device
CN117319847A (en) Multi-dimensional shaping combined coherent end-to-end equalization system
Elmeddeb et al. Robust Image Transmission Over Time-varying Channels using Hierarchical Joint Source Channel Coding
Pons et al. New tight upper-bounds on the error probability of multilevel coded modulation schemes
KR20090000998A (en) Apparatus and method for detection maximum likelihood of multiple input multipke output

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20953296

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2023512729

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20953296

Country of ref document: EP

Kind code of ref document: A1